Study on the role of protein deiminases in the nervous system

"Study on the role of protein deiminases in the nervous system" Citrullination (deimination) is the conversion of protein-bound arginine to citrulline. Citrullination is catalyzed by a family of calcium-dependent enzymes, the peptidylarginine deiminases (PADs). In mammals there are five PADs (PAD1, PAD2, PAD3, PAD4 and PAD6), whereas only 3 are found in chicken (PAD1-3). PAD isoenzymes are widely distributed in mammalian tissues and several studies suggest that citrullination occurs in extreme conditions such as during apoptosis, and during differentiation, when there is an increase in intracellular calcium concentration. Citrullination of different PAD target proteins has been associated with certain diseases, such as Alzheimer's disease, multiple sclerosis, rheumatoid arthritis and psorias. In humans, citrullination might be an early marker in neurodegenerative diseases. The role of PADs and citrullination in some human diseases is poorly understood and the physiological roles of PADs have yet to be fully investigated. At present citrullination is believed to play a role in myelin sheath formation and during keratinocytes terminal differentiation. Recent studies in mouse have suggested that PAD4 regulates histone methylation at the p21/WAF1/CIP1 promoters in a p53-dependent manner. The group of PF at UCL Institute of Child Health in London has identified PAD3 among calcium-dependent molecules differentially regulated in response to spinal cord injury at stages of development permissive (E11) and non-permissive (E15) for regeneration in chick embryos. Consistent with the up-regulation of PAD3 in spinal cords injured at E15, more extensive citrullination was observed after injury at this stage. This paralleled high apoptosis and significant tissues loss in injured E15 spinal cords. Following treatment at the time of injury with a PAD inhibitor, Cl-amidine, the secondary injury response in E15 spinal cord was greatly reduced. The aim of my project was first to see if it was possible to model the injury response in neural cells in vitro to eventually establish a model for studying PAD, citrullination and injury response in human neural cells and secondly I tried to better characterize expression pattern and possible role(s) of PAD, focusing mainly on PAD3, in neural cell death and survival initially using neural progenitor cells derived from E7 (embryonic day 7) chick midbrain (cMB cells), which were shown to express PAD3 both at the protein and mRNA level, and subsequently in a human neural tumor cell line, the neuroblastoma cell line LAN5. To this purposes we first studied the effects of PAD inhibition, using Cl-amidine, and PAD activation, using thapsigargin to raise intracellular calcium, on expression and cellular localization of PAD and on citrullination of a PAD target, histone 3 (H3) in cMB cells. In some experiments chick midbrain cells from E7 embryos were treated also with the p53 inhibitor, pifithrin, to assess whether p53 is required for PAD activity. Thapsigargin treatment induced significant cell death, but this was reduced by pretreating the cells with Cl-amidine. Reduction of cell death upon PAD inhibition parallels the in vivo finding in the injured chick spinal cord. Significantly, PAD3 mainly localized to the nucleus in cMB cells at passage 12 following thapsigargin treatment, whereas it remained largely cytoplasmatic in controls and cells pre-treated with Cl-amidine. Citrullinated H3 (CitH3) was not restricted to the nucleus in thapsigargin-treated cells while it is definetly nuclear in control and CL-amidine treated cells. These results suggest that the PAD-associated apoptotic effect is likely due to PAD activity in the nucleus and that increased citrullination of H3 results in its export from the nucleus. As cMB cells were found to become senescent and die around passage 30 (P30), we investigated whether expression of PAD3 and CitH3 changed with time in culture. At P30 PAD3 was largely nuclear both in control and treated cells (thapsigargin/Cl-amidine), and CitH3 distribution resembeld that observed at P12 following thapsigargin-treatment. No changes in PAD3 localization were observed upon thapsigargin/Cl-amidine-treatments at P30, unlike at P12 where PAD3 localized in the nucleus only in thapsigargine-treated cells. Western blotting showed no differences in the amount of CitH3 among the different treatments. These results suggest an increased nuclear activity of PAD3 and increased turnover of CitH3 with aging that may be associated with cellular senescence and death. As certain tumor cell lines were reported to express PADs, we assessed expression of PAD3 and PAD4, another PAD that can localize to the nucleus, and the response to thapsigargin and Cl-amidine treatment in LAN5 cells. PAD3 and PAD4 appeared to be mainly perinuclear, and their expression and localization was not affected by any drug combination (assessed by immunocytochemistry). Western blotting and immunocytochemistry showed that also CitH3 amount and localization were not affected in treated cells. In addition to an effect on cell death, PAD activity was found to play a role in cell adhesion. When plated in the presence of Cl-amidine, most cMB cells did not attach to the dish, but were able to do so upon removal of Cl-amidine. Therefore their adhesion ability rather than survival appear to be affected. When cells were grown in neural differentiation medium (serum-free) to study a possible role of PAD in differentiation, such recovery was not observed. Howewer, Given that PAD3 expression was observed both in cell positive and negative for the neuronal marker, ß3-tubulin, and that in the presence of serum Cl-amidine did not affect expression of ß3-tubulin, as indicated by immunocytochemistry and Western blotting, a role for PAD in neuronal differentiation does not seem to be likely. The expression pattern of PAD3 during chick embryonic development was assessed by mRNA analysis and RT-PCR in tissues dissected from embryos in variuos stage of development. Valentina Millart